123

My friend told me that he doesn't bother voting any more in elections, because "it's not like one vote ever made a difference." I will vote when I can and I think it is important, and although I share his sentiment about politicians... it got me thinking.

Has there ever been any major election (around 1 million votes or more) where a single vote has decided the result? e.g. Candidate A got 1,000,000 votes and Candidate B got 1,000,001 votes. If the voter for candidate B did not vote it would be a tie and there might be a run-off election or some other system used to determine the winner and thus B may have lost if that one vote was not cast.

Larian LeQuella
  • 44,977
  • 18
  • 187
  • 208
Thomas O
  • 11,908
  • 7
  • 53
  • 72
  • 1
    Depending on the political system, electorates could have only a few thousand voters in which case very tight finished are not uncommon. – Craig May 22 '11 at 12:01
  • 17
    Of course, even votes that *don't* decide the election "count" is a wider sense. When "minor" parties draw a non-trivial fraction of the votes, analysts know there is *something* on the electorate's mind. What happens then depends on the system, but it usually results in *somebody* adopting a new platform or slightly altering a stance in an effort to take advantage of that pool of voters. That's not much, but it is also more than nothing. – dmckee --- ex-moderator kitten May 22 '11 at 15:11
  • 7
    If a country has a system where parties submit slates and allocate representation proportionally, it would be more likely for a single vote to flip a representative from one party to another. – David Thornley May 22 '11 at 18:17
  • heard on the radio the other day where two local polititians had to draw straws since they tied. – Andy May 23 '11 at 07:36
  • Another good example that probably doesn't deserve its own answer is the [2008 Minnesota Senate Election](http://en.wikipedia.org/wiki/United_States_Senate_election_in_Minnesota,_2008#Recount) which was close enough to cause an automatic recount. Before the recount, Coleman was winning; Afterwards, Franken was winning. The final decision was for Franken by 312 votes. – MrHen May 23 '11 at 17:22
  • 15
    I'm surprised no one's mentioned the idea that really, _every vote counts no matter what_. This expands on what @dmckee said... imagine if two million people (and I would be surprised if this isn't true) took that stance in a US presidential election, that "oh, my vote won't matter." That's two million uncounted votes, right there. In an election of over one million votes, there's just statistically _always_ going to be more than one person not voting for a dumb reason, and it adds up. – erekalper May 27 '11 at 16:46
  • 1
    The point is, if you don't vote, and your friend isn't voting either, we are already talking about **two** votes. And because this attitude is quite common, it really does sum up. – Mackie Messer May 28 '11 at 00:53
  • 1
    I think the more salient point is that the margin of error in counting votes is likely to be more important than any single (or even a few hundred) votes. – JohnFx Jun 11 '11 at 03:04
  • 3
    It's possible that one-vote counts even in presidential elections. Keep in mind that America and many (most?) other western countries are not democracies, they're democratic-republics. This means your vote for president isn't 1/300,000,000th of the overall votes for president; it's more like 1/10,000th of the overall votes for your municipality. The majority of your municipality will decide how your ENTIRE municipality votes, and the majority municipality votes will decide how the ENTIRE state votes. This means in close elections one vote may very well have made a difference. – BlueRaja - Danny Pflughoeft Jun 16 '11 at 04:57
  • 25
    "No individual raindrop ever considers itself responsible for the flood." – JYelton Oct 14 '11 at 17:50
  • 22
    There's an unasked question here: Has the fallacious belief that "it's not like one vote ever made a difference" ever made a difference in the outcome of an election? If you can get 10% of your opposition to believe the above, then you've effectively raised your candidate's popularity by 10% or more among the only members of the electorate who matter, those who actually turn out to vote. –  Jan 14 '12 at 17:21
  • don't forget that 1 vote cast means a 2 vote difference in the outcome. 1 voter staying home means every other vote has more weight. – jwenting Oct 22 '12 at 04:46
  • Also worth noting: Your probability of determining the outcome of an election decreases as the *square root* of the number of voters (assuming votes are a random sample of a normal distribution, which isn't the case but may be close to the case in a close election). The importance of your vote is presumably directly proportional to the number of voters. So arguably you should want to vote more in a giant election even though you have little chance of swinging the outcome. – psr Nov 13 '12 at 18:41
  • 2
    If you were off by 1 vote then they would have a recount and you'd suddenly find you were off by 150 votes, so really 1 vote is easily lost in "error" alone. – Bill K Dec 30 '12 at 20:18
  • Individual breaths don't make much of a difference to your body, but you're lucky that you don't breathe at the same rate that people have recently voted in UK elections. – Paul D. Waite Sep 12 '13 at 10:48
  • Well, after counting and recounting and error checking, Obama won over Hillary in my area to get on the presidential ballot by 1 vote. 1 can certainly matter, even if it's not common! – Brian S Nov 18 '13 at 17:25
  • This is now been sufficiently answered as a predicate (instances do exist) but is now being treated as a "list question" which is frowned upon in StackExchange land. – Oddthinking Nov 19 '13 at 02:55
  • 1
    According to https://en.wikipedia.org/wiki/List_of_United_States_congressional_districts the average congrestional district in USA has 710000 voters. So your 1 million voters requirement sets a fairly high bar. – Taemyr Jul 06 '15 at 13:08
  • Next vote will make a difference if your gold badge count - so Yes :) – Matas Vaitkevicius Jan 12 '16 at 17:08
  • I think it's more accurate to say that the voting population's *belief* in whether or not one vote makes a difference is what makes the biggest difference. – Jason C Mar 09 '16 at 03:10
  • I apparently can't post a new answer, but in August 2015 there was a referendum planned in Columbia, Missouri with only one registered voter: http://www.columbiatribune.com/news/local/college-student-would-be-sole-voter-in-cid-sales-tax/article_6702c44b-0243-51f8-861c-1df0b462cd92.html – Eric May 14 '16 at 23:02

13 Answers13

138

Answer: Yes!

Although they weren't major. Örebro (Sweden) has around 100,000 inhabitants.

In local municipality elections in Sweden, there have been cases of one vote making a difference in determining which party gets a seat. In the election of 2010, this single vote difference in determining the last seat of Örebro municipality actually meant that the socialist block got the majority there.

Of course, every vote counts. And every vote makes a difference, so it wasn't one vote that made a difference it was all the votes that made a difference. Every single one of the votes for the socialist block gave the victory to that block, because without just one of them, it would have been a lottery. (Literally, they would have had a tombola draw.)

Ref: En röst avgjorde valet i Örebro, SvD (English translation).

And there are other cases of this in Sweden. In a referendum in 1971 there was only one vote's difference. En enda röst avgjorde när Gamleby bestämde sig, vt.se (English translation)

Giacomo1968
  • 5,508
  • 6
  • 25
  • 52
Lennart Regebro
  • 8,919
  • 1
  • 47
  • 43
  • 39
    +1. Every vote counts - I learned that when becoming the president thing of my highschool, where I basically won by one vote :D.... unfortunately, ever since that day, every single person who voted for me kept annoying me saying *"it was MY vote, so you owe me one..."* – Saturn Mar 06 '13 at 16:23
  • 3
    Since the original question asked about elections of over a million votes and the accepted answer refers to a town of 100,000 people the answer is actually a clear "no" – Bill K Jul 06 '15 at 14:19
  • @Dr.Nobody No, it doesn't count. Having a small voter group of say.. 100 people (your highschool) is nowhere near elections with millions of voters. With 100 votes its not that hard to get a tie. With millions.. it's extremely rare. – user Oct 01 '15 at 12:39
  • 2
    A new example today in Virginia, where a House of Delegates race came down to a single vote, 11,607 to 11,608. The result is even more significant, because it means the House of Delegates is now divided 50-50. – Mark Dec 20 '17 at 01:39
  • And then it was actually 11,608 to 11,608. :-) Definitely a case where all votes count. – Lennart Regebro Jan 12 '18 at 07:06
60

The Danish Parliamentary election in 1998 was decided by 179 Faroese votes out of about 2 million votes.

Not a single vote, but close enough that that year every vote counted. I believe that percentage-wise it was even closer than the Bush-Gore election in 2000.

Reference: Nyrup vandt nervepirrende valg ("Nyrup won the nerve-wracking election" in Danish)

Laurel
  • 30,040
  • 9
  • 132
  • 118
Thorbjørn Ravn Andersen
  • 718
  • 4
  • 7
  • 7
    -1, since this does not answer the question. 179 > 1. –  May 24 '11 at 13:15
  • 41
    @Tim: This answer is helping establish a baseline for the question. "1" is a number. After arbitrarily choosing 1000000:1 as the target, this answer hits 2000000:179. While it doesn't _technically_ answer the question, is it not being helpful? – MrHen May 25 '11 at 13:34
  • 3
    @MrHen: I see the target as exactly 1 rather than 10^6:1. A world-wide referendum where 6000 votes make the difference would not answer the question. By the way Torbjorn, don't worry, I don't have enough rep to downvote :) –  May 25 '11 at 13:39
  • 2
    @Tim: Sure. The point I was making is that the "target" may not the only useful thing to say on this subject. – MrHen May 25 '11 at 13:45
  • That someone needs to go here kind of answers the question, NO one vote never made a difference, eh? 179 is no where NEAR 1 when you think about the fact that if it was off by 2 you'd still be no where near being off by 1 – Bill K Dec 30 '12 at 20:16
  • @BillK please see http://skeptics.stackexchange.com/a/3519/3021 then. – Thorbjørn Ravn Andersen Dec 30 '12 at 21:34
  • @BillK then it would perhaps be a good idea to introduce the concept of "large election" (and what determines when you have one) into the original question. – Thorbjørn Ravn Andersen Jul 06 '15 at 10:33
  • 2
    The original question asked about elections over a million people. The one in your comment link above to skeptics was much smaller. – Bill K Jul 06 '15 at 14:24
43

Wikipedia has a list of them at List of close election results.

It lists several cases where the vote was tied, or had margins of one or more vote, but most of the cases it lists are individual seats being decided by a single vote, rather than the balance of power in parliament/congress.

The only vote with wider significance would be the Massachusetts gubernatorial election, 1839. Wikipedia claims that Marcus Morton won by two votes, which would mean that one person could have changed the margin from 2 votes to a dead heat. However, it doesn't provide a citation.

Andrew Grimm
  • 38,859
  • 36
  • 141
  • 342
22

The book "History of the Indiana Democracy 1816-1916" written by John B. Stoll in 1917, describes the outcome of an Indiana state senate election in 1841:

Dr. Madison Marsh, noted Democrat of the county, lost on the face of the returns by one vote to his Republican opponent. Captain Beall. On a contest it was decided that one vote for Mr. Marsh, that of Henry Shoemaker, whose ballot had been received after the time limit but before the box was closed, had been improperly thrown out and he was declared elected to the Legislature. Edward A. Hannegan was elected United States Senator by one vote, Dr. Marsh casting the deciding vote. Texas was admitted to the Union, it is declared, by a majority of one, Hannegan being (he [sic] man of destiny to decide the momentous issue.

The Texas State Library and Archives Commission confirms that one step of the annexation process passed 27-25 in the US Senate. Of course history doesn't record whether Mr. Shoemaker himself supported the annexation of Texas. A Hoosier in the 1840s might have had mixed feelings about adding another southern slave state to the Union. But his one vote evidently did "make a difference".

NonSequitur
  • 1,219
  • 10
  • 11
  • 3
    Note: there were less than 1 million people in the entire state of Indiana in 1841 so this answer does not strictly meet your criteria. – NonSequitur May 22 '11 at 13:42
  • 6
    Which brings up a good point. At the time this myth was popularized, I doubt there were 1 million people voting anywhere at all, let alone in an election decided by one vote. – MrHen May 25 '11 at 13:36
18

According to Repubblica, it just happened for the mayor of Meda (MB, Italy), won by 3867 vs. 3866.

enter image description heresource

Sklivvz
  • 78,578
  • 29
  • 321
  • 428
o0'.
  • 1,307
  • 2
  • 10
  • 14
  • 1
    The question is about major elections (millions of voters). This answer is about a small election of 7,000 voters. – user Oct 01 '15 at 12:31
11

One more win by a margin of 1 vote :

The Times of India - First to lose elections in Karnataka by one vote says,

In the studies conducted by American psephologists in the history of American democracy of 16,577 national elections from 1898 , they have found that only one's electoral fate had been decided by a single vote. It was in the 1910 election in New York's 36th Congressional District a Democrat who polled 20,685 votes to the Republican candidate's 20,684 was declared elected by one vote .

In India this one vote victories became a reality with the advent of electronic voting machines where there is no scope for human intervention and error of judgment in deciding the votes and Karnataka became the first state to record such an incident . The first candidate to enter the book of records is none other than the son of veteran politician and former governor of Kerala late B Rachiah's son A R Krishnamurthy who lost against R Dhruvanarayan in Santhemarahalli(Sc) constituency in the 2004 assembly elections. While Krishnamurthy contesting on JDS ticket polled 40751 against Dhruvanarayan's 40752 votes .

Community
  • 1
Sumanth
  • 117
  • 1
  • 2
  • 1
    The question asks specifically about elections with millions of voters. 40,000 voters are much less than 1,000,000. – user Sep 30 '15 at 21:08
10

Stephen Dubner and Steven Levitt, the economists behind Freakonimics tackled this topic in an article for the NY Times about 5 years ago. It's a good read.

Excerpt:

The odds that your vote will actually affect the outcome of a given election are very, very, very slim. This was documented by the economists Casey Mulligan and Charles Hunter, who analyzed more than 56,000 Congressional and state-legislative elections since 1898. For all the attention paid in the media to close elections, it turns out that they are exceedingly rare. The median margin of victory in the Congressional elections was 22 percent; in the state-legislature elections, it was 25 percent. Even in the closest elections, it is almost never the case that a single vote is pivotal. Of the more than 40,000 elections for state legislator that Mulligan and Hunter analyzed, comprising nearly 1 billion votes, only 7 elections were decided by a single vote, with 2 others tied. Of the more than 16,000 Congressional elections, in which many more people vote, only one election in the past 100 years - a 1910 race in Buffalo - was decided by a single vote.

But there is a more important point: the closer an election is, the more likely that its outcome will be taken out of the voters' hands - most vividly exemplified, of course, by the 2000 presidential race. It is true that the outcome of that election came down to a handful of voters; but their names were Kennedy, O'Connor, Rehnquist, Scalia and Thomas. And it was only the votes they cast while wearing their robes that mattered, not the ones they may have cast in their home precincts.

For the record, I think that last bit about the Bush v. Gore Supreme Court decision is obscenely oversimplified and in my personal opinion an inaccurate characterization of what happened.

JohnFx
  • 1,145
  • 1
  • 10
  • 19
  • 2
    RE Bush-Gore 2000: Yeah. If the Supreme Court had said, "Candidate X won by 10,000 votes, but we think candidate Y would make a better president so we're going to ignore those votes", that would be very different. The issue was never whether the votes of the people mattered, but arguments over how to accurately determine the vote and how to deal with ballots that were not filled out correctly, like someone voting for 2 candidates when they were supposed to pick 1. – Mark Daniel Johansen Dec 28 '18 at 14:55
8

When I was young (8), I remember our city's mayoral election being decided by one vote in Kenosha, Wisconsin (USA). The final tally was 13,114 to 13,113. Here is a newspaper article, courtesy of Google News: http://news.google.com/newspapers?nid=1368&dat=19800409&id=KnpQAAAAIBAJ&sjid=AxIEAAAAIBAJ&pg=5271,1364207

Joel Rondeau
  • 189
  • 2
  • 5
  • 1
    The question asks about 1 vote out of million of voters. 26,000 is far from a million. – user Jul 05 '15 at 09:13
7

Most electoral systems have provisions in case of ties, though with large amount of voters this is rare in public elections. Tiebreakers usually take place in form of coin toss or some similar game of chance. Ties are (comparatively) common in legislative bodies.

The article King Trumps 5 as Card Flip Settles an Election Tie in North Las Vegas describes such an election in Nevada where the election result was decided through a tiebreaker (card draw). Similarly an election in the UK was decided by a coin toss.

In a personal anecdote, in my hostel elections once the winning candidate won by one or two votes. My roommate had cited personal inconvenience and the above mentioned "one vote does not make a difference" and decided not to vote, but was persuaded by people to vote. He always claimed credit for the winning vote later.

Laurel
  • 30,040
  • 9
  • 132
  • 118
apoorv020
  • 3,266
  • 2
  • 25
  • 37
  • 1
    There was one of these in south-east New Mexico while I lived there for a seat in the State House of Reps. The law in those parts allow for ties to be decided by *"a mutually acceptable game of chance"*. But it wasn't a million people. – dmckee --- ex-moderator kitten May 22 '11 at 15:07
4

This municipal vote on whether to fund the building of a new library initially came out in a 522-522 vote tie until a recount reached a 523-522 result.

The original link above is broken. http://www.rigorousintuition.ca/board2/viewtopic.php?f=34&t=33947 works at the moment, and is a reproduction of the article titled, "Once again, tie vote defeats new library in Shutesbury".

ChrisW
  • 26,552
  • 5
  • 108
  • 141
BlueWhale
  • 931
  • 7
  • 14
  • The question asks about 1 vote out of million of voters. This answer is about 500 voters. – user Sep 30 '15 at 21:10
3

In the recent Landtagswahl (state-wide election) in Kärnten, Austria, one vote took the mandate of the city St. Veit an der Glan away from the party "BZÖ" and gave it to the "Grüne".

The very ballot paper, according to this report, was counted valid even though the candidate of the Green party was not marked by a cross or a check mark, but by a crudely drawn, not-safe-for-minors sketch referring to a certain part of the male anatomy.

doppelfish
  • 131
  • 3
2

A Radio Lab episode was recently published on this very topic.

They debunk a reader's letter to Ann Landers - see Snopes for more.

They retell an interesting tale about an Indian election that was initially lost by a single vote. As Mental Floss describe it:

In 2008, an Indian politician named C.P. Joshi lost by a single vote pursuing an assembly position in the North West Indian state of Rajasthan. In the final tally, Joshi fell to opponent Kalyan Singh Chouhan by a count of 62,216 to 62,215.

The dramatic twist? The winner's wife was then investigated for voting twice, and the result was declared void.

Oddthinking
  • 140,378
  • 46
  • 548
  • 638
1

The answer depends on what "difference" means in the question text. As you further elaborated, it can mean a very close election decided by just one vote. Of course, statistically and factually this is rare but it is probable and other users gave some examples. The important thing is the vote that made the difference in such an election is not a particular person's vote. For instance, in an election that resulted in 100000 to 999999 all the 1000000 votes cast is responsible of the election result not just 1 particular vote. Without the other 999999 similar votes such a vote would mean nothing.

So election result is an emergent property of all the votes cast. As some commenters said, an election result is also not just about who the winner is or how close it is. If that was the case there would be no analysis of elections in newspapers or academia. They will just report the results.

So if you interpret "difference" in another sense, every vote makes a difference. You can read about the following topics in Wikipedia for further examples of this phenomena:

pembeci
  • 143
  • 1